DECIDE · STEP 1 OF 4
The readiness checklist
Before comparing vendors or frameworks, answer six questions about your data. They determine whether any MMM, at any price, can produce estimates you would defend in front of a CFO. Answer honestly; the model will find out anyway.
By Oliver Wakefield-Smith · Updated 2 August 2026 · Every number on this page resolves to a primary source in the ledger below.
Do you have at least 104 weeks (2 years) of weekly revenue and spend history?
Do you spend meaningfully in at least 4 distinct channels?
Has spend actually varied (launches, pauses, seasonality), rather than holding flat?
Is there one revenue source of truth (finance-grade, not platform-reported)?
Can you split revenue and spend by region or market?
Do you record non-marketing drivers (price changes, promos, stockouts, PR spikes)?
Why these six
They map to the failure modes documented across the major frameworks: history and seasonality separation, channel breadth, spend variation, dependent-variable hygiene, geo granularity, and confounder logging.12 Full detail on each lives in data requirements and MMM limitations.
Failing checks is normal; most brands fail two on first pass. The fixes are listed under each "No", and most cost weeks, not quarters. Failing four or more is a different signal: read when not to use MMM before any vendor call, and let the checklist verdict above say so to your stakeholders in writing.
Passed? Continue to step 2: build vs buy.
Source ledger
Every numeric claim on this page resolves to one of these primary sources.
- [1]Google Meridian documentationhttps://developers.google.com/meridianretrieved 2026-08-02
- [2]Meta Robyn feature documentationhttps://facebookexperimental.github.io/Robyn/docs/featuresretrieved 2026-08-02